HOTD

Proposed a holistic cross-layer time delay attack detection framework (HOTD) to detect time delay attack in UAV networks, which is easy to implement and difficult to detect. First, we perform a holistic selection of delay-related features at each layer of UAV networks. Then, supervised learning is used to construct a consistency model between these selected features and the corresponding forwarding delay, based on which the consistent degree of each node can be calculated. Finally, the K-Means clustering method is utilized to distinguish malicious nodes from benign ones according to their consistent degrees. (2021-2022, published in journal Journal of Parallel and Distributed Computing (JPDC) [CORE A, CCF B, SCI-Q1, IF 4.542])